Biological Journal of the Linnean Society, 2020, XX, 1–10. With 2 figures.
Trade-offs between burrowing and biting force in fossorial scincid lizards?
MARGOT LE GUILLOUX1, AURÉLIEN MIRALLES2, JOHN MEASEY3, BIEKE VANHOOYDONCK4, JAMES C. O’REILLY5, AURÉLIEN LOWIE6, and
ANTHONY HERREL1,4,6,
*
1UMR 7179 C.N.R.S/M.N.H.N., Département Adaptations du Vivant, Bâtiment d’Anatomie Comparée, 55 rue Buffon, 75005, Paris, France 2Institut de Systématique, Evolution, Biodiversité, (UMR 7205 Muséum national d’Histoire naturelle, CNRS UPMC EPHE, Sorbonne Universités), CP30, 25 rue Cuvier 75005, Paris, France 3Centre for Invasion Biology, Department of Botany & Zoology, Stellenbosch University, Private Bag X1, 7602 Matieland, Stellenbosch, South Africa 4Deparment of Biology, University of Antwerp, Universiteitsplein 1, B2610 Antwerpen, Belgium 5Department of Biomedical Sciences, Ohio University, Cleveland Campus, SPS-334C, Cleveland, 45701 Ohio, USA 6Department of Biology, Evolutionary Morphology of Vertebrates, Ghent University, K.L. Ledeganckstraat 35, 9000 Ghent, Belgium
Received 27 December 2019; revised 17 February 2020; accepted for publication 19 February 2020
Trade-offs are thought to be important in constraining evolutionary divergence as they may limit phenotypic diversification. The cranial system plays a vital role in many functions including defensive, territorial, predatory and feeding behaviours in addition to housing the brain and sensory systems. Consequently, the morphology of the cranial system is affected by a combination of selective pressures that may induce functional trade-offs. Limbless, head-first burrowers are thought to be constrained in their cranial morphology as narrow heads may provide a functional advantage for burrowing. However, having a wide and large head is likely beneficial in terms of bite performance. We used 15 skink species to test for the existence of trade-offs between maximal push and bite forces, and explored the patterns of covariation between external head and body morphology and performance. Our data show that there is no evidence of a trade-off between bite and burrowing in terms of maximal force. Species that generate high push forces also generate high bite forces. Our data also show that overall head size covaries with both performance traits. However, future studies exploring trade-offs between force and speed or the energetic cost of burrowing may reveal other trade-offs.
ADDITIONAL KEYWORDS: covariation – cranial system – divergence – head-first burrowers – morphology – skink.
anatomical structures (Arnold,1992;Van Damme et al., 2002, 2003). Previous studies have further suggested that trade-offs may in some cases limit phenotypic variation by constraining evolutionary divergence
(Vanhooydonck et al., 2001; Levinton & Allen, 2005; Konuma & Chiba, 2007; Herrel et al., 2009). The
cranial system plays a vital role in many activities including defensive, territorial, predatory and feeding behaviours in addition to housing and protecting the brain and major sensory organs (Andrews et al., 1987;
Cooper & Vitt, 1993; Herrel et al., 2007; Kohlsdorf
INTRODUCTION
The phenotype of an organism reflects the selective pressures exerted by activities that are essential to its survival and its reproduction (Arnold, 1993). Sometimes, however, the functional demands exerted by different performance traits may result in tradeoffs. Indeed, functional trade-offs arise when different functions exert conflicting pressures on the same
*Corresponding author. E-mail: anthony.[email protected] © 2020 The Linnean Society of London, Biological Journal of the Linnean Society, 2020, XX, 1–10
1
2
Burrowing and Biting in Scincid Lizards
et al., 2008; Dumont et al., 2016). Consequently, the
morphology of the cranial system is affected by these combined selective pressures which may potentially induce functional trade-offs. burrowing force. Consequently, trade-offs between these two performance traits, if present, are not mediated by direct functional conflict for the optimization of single functional trait. Rather, burrowers can be expected to have narrow heads for efficient soil penetration which may come at a cost of bite force.
The hypothetical trade-off between biting and borrowing performance is particularly interesting in limbless burrowing animals. These organisms are obligate head-first burrowers and the evolution of their morphology is thought to be constrained. Indeed, because the energy required for burrowing increases exponentially with body and head diameter (Navas et al., 2004), having a thin body and a narrow head may provide a functional advantage for burrowing.Yet, this is likely detrimental in terms of bite performance
(Verwaijen et al., 2002; Navas et al., 2004; Herrel & O’Reilly, 2006; Vanhooydonck et al., 2011; Baeckens
et al., 2017; Hohl et al., 2017). Maximum bite force has
been suggested to limit the type and size of food an
animal can eat (Herrel et al., 1999, 2001, 2008; Aguirre et al., 2003; McBrayer & Corbin, 2007; Edwards et al.,
2013). Consequently, a cranial form optimized for soil penetration may compromise the types of food an animal can eat and vice versa (Andrews et al., 1987;
Barros et al., 2011; Baeckens et al., 2017).
Here, we test for the presence of a trade-off between maximal bite force and the maximal push force in a range of burrowing and leaf-litter dwelling skinks. Scincid lizards represent an ideal study system as this family includes a variety of ecologies and morphologies with quadrupedal surface-dwelling species, epigeal serpentiform species with partially reduced front- and/ or hindlimbs, burrowing completely limbless species, and a plethora of intermediate forms (Pianka & Vitt, 2003;
Miralles et al., 2015; Wagner et al., 2018; Bergmann &
Morinaga, 2019). At least 15 independent evolutions of a serpentiformbodyformhavetakenplacewithinthegroup (Benesh & Withers, 2002; Miralles et al., 2012) allowing for a robust framework to test for associations between life-style, performance and morphology. Consequently, we also explore the patterns of covariation between head and body morphology and the two performance traits studied here (bite force and push force).
Burrowing is a complex behaviour that remains rather poorly understood in limbless head-first burrowing vertebrates (but see for example Gaymer
(1971); Gans (1973); O’Reilly et al. (1997); Teodecki et al. (1998); Navas et al. (2004); De Schepper et al. (2005)).
The maximal push force that an animal can generate is likely important as it may allow an animal to penetrate a greater variety of soil types, and thus expand its resource base in terms of potential habitat and food resources. As limbless species burrow by recruiting muscles along the long axis of the body (Rieppel, 1981;
Navas et al., 2004; Vanhooydonck et al., 2011; Hohl
et al., 2017), the diameter and size of the body should be related to the maximal push force it can generate. However, to facilitate soil penetration the width of the head should rather be narrow as this optimizes the pressure for a given force (e.g. Measey & Herrel, 2006;
Herrel & Measey, 2010; Barros et al., 2011). Moreover,
the speed by which it can penetrate the soil (and not only the force generated) is also likely a factor significantly contributing to the burrowing performance (Ducey
et al., 1993; Teodecki et al., 1998; Vanhooydonck et al.,
2011). Although few quantitative data exist, a previous study suggested the presence of a trade-off between bite force and burrowing speed in a limbless skink, Acontias percevali, mediated by the conflicting demands on head dimensions (Vanhooydonck et al., 2011). However, whether this is more generally the case and whether trade-offs also exist between bite force and push force remains unknown. As burrowing force is dependent on the axial musculature, different anatomical traits are responsible for the generation of bite force vs.
MATERIALS AND METHODS
AnimAls
Morphological measurements were performed on 180 individuals and performance measurements were obtained for 171 individuals for bite forces and 159 for push forces across 14 different species (Table 1).Animals were sampled between 2000 and 2017. The specimens were adults of often unknown sex. Data were collected in situ in the field or in the lab for species that were obtained through the pet trade. An additional five individuals of the species Pygomeles braconnieri from the collections of the National Museum of Natural History in Paris were used for morphological measurements.
morphometrics
Each individual was weighed using an electronic balance (Ohaus, 0.1 g). Head length, head width, head depth and lower jaw length were measured using a digital caliper (Mitutoyo, 0.01 mm) as described previously (Herrel and Holanova, 2008). The snoutvent length was measured by stretching the animals along a ruler ( 1 mm).
mAximAl push force
Maximalpushforcesweremeasuredinthefieldorinthe lab following the protocol described in Vanhooydonck
© 2020 The Linnean Society of London, Biological Journal of the Linnean Society, 2020, XX, 1–10
M. LE GUILLOUX ET AL.
3
© 2020 The Linnean Society of London, Biological Journal of the Linnean Society, 2020, XX, 1–10
4
Burrowing and Biting in Scincid Lizards
Figure 1. Phylogeny used in the analyses modified from Pyron et al. (2013).
et al. (2011). Measurements of peak push forces during
mAximAl bite force
burrowing were made using a custom piezoelectric
Maximal bite forces were measured in the field or
the lab following Herrel et al. (1999). In brief, in vivo bite forces were measured using an isometric Kistler force transducer (type 9203, Kistler Inc.) mounted on a purpose-built holder and connected to a Kistler charge amplifier (type 5058 A, Kistler Inc.). Biting causes the upper plate to pivot around the fulcrum, and thus pull is exerted on the transducer. Capture of the animals typically resulted in a characteristic threat response where the jaws are opened maximally. The free end of the holder was then placed between the jaws of the animal, which immediately resulted in fierce and prolonged biting. The gape angle ( 30 °) and the place of application of the bite force was standardized with animals always biting at the tips of the jaws. Measurements were repeated five times for each animal and the maximum value recorded was considered to be the maximal bite force for that animal. force platform (Kistler Squirrel force plate, 0.1 N, Kistler Inc., Switzerland). The force platform was positioned on a custom-designed metal base and connected to a charge amplifier (Kistler Charge Amplifier type 9865, Kistler Inc.).A Perspex block with 1 cm-deep holes of different diameters was mounted on the force plate, level with the front edge. One of the holes was loosely filled with soil from the container of the animal that was tested. A Perspex tunnel with a diameter similar to the maximal body diameter of the test animal was mounted on the metal base in front of (but not touching) the force plate, and aligned with the soil-filled hole in the Perspex block. First, a skink was introduced into the tunnel and allowed to move through it until reaching the soil-filled chamber. Next, the animal was stimulated to burrow into the soil by tapping the end of the tail sticking out of the tunnel, or by prodding the animal inside the tunnel with the blunt end of a thin wooden stick. Forces were recorded during a 60 s recording session at 500 Hz, and three trials were performed for each individual, with at least 1 h between trials. Forces were recorded in three dimensions using the Bioware software (Kistler Inc.). For each individual we then extracted the highest peak resultant force across all trials as an indicator of that animal’s maximal push force.
AnAlyses
- Morphometric and force data were Log10
- -
transformed before analysis to ensure normality and homoscedasticity. All analyses were performed in R (v.3.4.0) while taking into account the phylogenetic relationships among species. The phylogenetic framework
© 2020 The Linnean Society of London, Biological Journal of the Linnean Society, 2020, XX, 1–10
M. LE GUILLOUX ET AL.
5
Figure 2. Phylogenetic regressions of species averages of maximal bite force against maximal push force. (A) Absolute push force is significantly correlated with absolute bite force. (B) Residual bite force is significantly and positively correlated with residual push force when correcting for overall snout-vent length. (C) Residual bite force is no longer related to variation in residual push force when corrected for variation in body mass.
© 2020 The Linnean Society of London, Biological Journal of the Linnean Society, 2020, XX, 1–10
6
Burrowing and Biting in Scincid Lizards
Figure 3. Covariance between morphological and performance data. (A) Phylogenetic two-block partial least square analysis between morphology and performance illustrating that more robust species (i.e. with greater body mass and bigger heads) produce greater push and bite forces. (B) Two-block partial least square analysis on the force and morphological data corrected for variation in snout-vent length. Species with relatively higher mass, heads size and body diameter produce greater bite and push forces. (C) Two-block partial least square analysis on the force and morphological data corrected for variation in body mass showing that species that are relatively more elongate (higher snout-vent length for their mass) show relatively higher push forces. In contrast, species that are stockier, less elongate and with bigger heads produce relatively higher bite forces.
used was obtained from the molecular data set of Pyron et al. (2013); Fig. 1. This required the reconstruction of a phylogeny by pruning the tree to include only species for which we had performance and morphological data. First, a phylogenetic generalized least squares (PGLS) regression was
© 2020 The Linnean Society of London, Biological Journal of the Linnean Society, 2020, XX, 1–10
M. LE GUILLOUX ET AL.
7
run in R (function gls with corBrownian in Phytools (Revell, 2012)) with the mean maximal push force per species against mean maximal bite forces to test for the presence of a trade-off between bite and push force. To test for co-variation between morphology and performance we ran a phylogenetic two-block partial least squares analysis (φ2bPLS) using the function ‘phylo.integration’ implemented in ‘geomorph’ in R (Adams et al., 2013). This method quantifies the degree of association between two data matrices. It is a descriptive multivariate analysis robust to multicollinearity between variables and therefore suitable for the use of morphometric and performance variables. The analysis generates axes that explain the covariance between the two data matrices. As body size simultaneously impacts head and body dimensions and forces, we ran PGLS analyses with snout-vent length (SVL) or body mass as our predictor and morphometric and performance traits as our independent variables and extracted the unstandardized residuals. Next, we used a Pearson correlation test to explore the existence of a tradeoff between our two residual performance traits independent of variation in overall body size. Finally, we ran a two-block partial least squares analyses (2BPLS) on the residual data to explore patterns of covariation between morphology and performance independent of variation in overall size using‘two.b.pls’ script of the ‘geomorph’ package (Adams et al., 2013), and ‘pls2B’ script of the ‘Morpho’ package (Schlager, 2013) in R. appear to have relatively large bite forces but low push forces.
covAriAtion between morphology And performAnce
The phylogenetic 2BPLS analysis was significant (rPLS = 0.983, P = 0.001). Heavier, larger and wider animals with longer, taller and wider heads produce larger forces (Fig. 3A). The 2BPLS analysis run on the size-corrected data was also significant (rPLS = 0.969, P < 0.001; Fig. 3B).This analysis indicates that the maximal push force and the maximal bite force co-vary, both principally with overall head and body robustness, with animals that are more elongate producing relatively lower push and bite forces. An analysis on the size corrected data using body mass was also significant (rPLS = 0.73, P = 0.012), yet showed a different pattern. This analysis shows that more elongate animals produce relatively higher push forces whereas the more robust limbed species with wide heads and bodies produce low push but high bite forces for their body mass (Fig. 3C).
DISCUSSION
Our results, based on a broad range of burrowing and leaf-litter dwelling skinks, show that there is no direct trade-off between bite force and burrowing force in this group. Species that produce strong push forces are also those which produce strong bite forces in both absolute and relative terms. These results suggest that whereas bite force may trade-off with burrowing
speed (Teodecki et al., 1998; Vanhooydonck et al., 2011)
this may not be the case for push force. Indeed, the same traits that favour increased bite force (large, robust heads and wide bodies) also appear to favour high push forces, at least in absolute terms. This makes intuitive sense as the muscles used to generate both bite and push force are positioned to the lateral side of the body in scincid lizards (Huyghe et al., 2009; Vanhooydonck et al., 2011). For example, the external adductor muscle is one of the largest contributors to overall bite force generation and lies external to the mandible (Vanhooydonck et al., 2011). Similarly, the iliocostalis and longissimus muscles that generate the push forces measured are also positioned laterally to the vertebral column. Consequently, wider heads and wider bodies should induce an increase in both absolute bite and push force. However, whereas these traits may indeed favour absolute force, the speed by which animals can penetrate the soil may be negatively impacted by the presence of wider heads
and bodies (Teodecki et al., 1998; Vanhooydonck et al.,
RESULTS
trAde-offs between bite force And push force
The linear regression taking into account phylogeny shows that maximum push force is positively correlated with maximal bite force (PGLS: r = 1.36, P < 0.001), suggesting that species that produce strong push forces are also those who produce strong bite forces (Fig. 2A). Analyses performed on the snoutvent length corrected data show a similar result (r = 0.89, P < 0.001) with animals that bite harder for a given size also being better pushers (Fig. 2B). However, when correcting force measurements for body mass, residual bite force was no longer correlated with residual push force (r = -0.63, P = 0.053; Fig. 2C). An inspection of the plot (Fig. 2C) suggests that more elongate and smaller species like Typhlosaurus
vermis and T. l omiae as well as Acontias litoralis
and A. kgalagadi tend to have relatively higher push forces whereas the stockier, larger species like
Scincus scincus, A. meleagris, and Mochlus sundevalli
© 2020 The Linnean Society of London, Biological Journal of the Linnean Society, 2020, XX, 1–10
8
Burrowing and Biting in Scincid Lizards
2011). Moreover, the energetic cost of burrowing may also be negatively impacted by these traits (Navas et al., 2004). However, whereas the limbed species are the best biters in absolute terms and when corrected for body mass the more elongate specialized burrowers
(e.g. Acontias, Scelotes, Typhlosaurus) actually produce
greater push forces. Indeed, the specialized head-first
limbless burrowers like Acontias or Typhlosaurus
have higher push forces for their mass despite being more elongate and less robust. Moreover, individuals of these species were usually captured at deeper soil depths and in less sandy soils (A. Herrel & J. Measey, pers. obs.) suggesting that they are overall better at burrowing. Thus, rather than absolute push force relative push force may be the principal driver of burrowing specialization. However, given the paucity of quantitative data on soil hardness and burrow depth in fossorial animals a quantitative analysis of these patterns is not possible. Additionally, it would be of interest to gather similar data on other groups of fossorial skinks from different radiations (Australia, Madagascar, the Philippines, see Wagner et al. (2018);
Morinaga & Bergmann, 2020) as the data in the
present study are strongly biased towards African skinks, especially the Acontinae.
Vanhooydonck et al., 2011; Hohl et al., 2017) as also
suggested by our data. Indeed, more elongate species of the genera Typhlosaurus as well as the more elongate
Acontias species (A. litoralis, A. kgalagadi) showed
high push forces yet low bite forces (Fig. 3C). It would consequently be of interest to perform finer scale analyses of head shape or skull shape using geometric morphometric approaches. This may also allow teasing apart of differences between the back and the front of the skull likely impacted by constraints on biting vs.
burrowing (Cornette et al., 2015).